Application Of Fuzzy Logical Control In Heat Exchanger System

Heat exchangers are standard components in the field of Power Engineering, and also other process industry. Temperature control is still a major challenge if the heat exchanger is operated over a wide range. The nonlinear behavior depends strongly on the flow rates and on the temperature of the media. In this paper, a cross-flow steam/water heat exchanger is considered. The goal is to control the water outlet temperature by manipulating the flow rate of the steam and control the water inlet rate by manipulating the height of the water level in the water tank.Conventional PID-type controllers have played a dominant role in the heat exchanger systems due to its simple control structure and ease of design. However, PID-type controllers may encounter problems such as lager overshoot and even poor stability in the control of plants with time delay.In recent year, plenty of advanced control techniques have been developed. Fuzzy Logical Control is one of these new paradigms and has become very attractive in industrial practices. Because of its capability to emulate human thinking and human inference and to capture the inexact information of the plant, Fuzzy Logical Controller (FLC) often yields satisfactory performances. In most FLCS, the universes of discourse of the error and the change of the error and the membership function of their fuzzy sets have to be determined in advance. Usually, this is not easy and requires the trial and error method.This paper designs a cross-flow steam/water heat exchanger and the according system and the sampling card. In the water inlet rate control, a PID-Fuzzy two-mode controller is used. In the outlet temperature control, this paper proposes to use Takagi-Sigeno Type Fuzzy System to construct controller for plants with time delay. The developed digital controller in this way is very simple because it has a compact analytical expression and includes only a few adjustable parameters. To improve the response speed, a P controller is added in the initial stages. So a P-Fuzzy two-mode controller bring out. Simulation results illustrate that the proposed controllers are effective in suppressing the maximum overshoot and in reducing the decaying oscillation cycles and also superior then PID controllers.